Question

A doctor wants to determine if smoker status, whether or not someone consumes alcohol, height and weight are good predictors of life expectancy. Using the data below perform the LINEST function in Excel. Do the overall hypothesis test and then test each individual slope. Which variables were not good predictors? Write out your final equations.

Age	Smoker	Alcohol Consumption	Weight (in lbs)	Height (in inches)
41	2	2	185	65
40	2	2	185	65
25	2	2	185	65
42	2	1	153	63
41	2	2	248	64
35	2	1	185	67
39	2	2	120	60
25	1	1	135	62
20	1	1	140	67
44	2	1	165	66
26	1	2	150	66
32	1	2	110	61
35	2	2	105	60
48	2	2	190	69
45	2	1	150	67

Please note: 1 indicates yes, 2 indicates no.

Answer 1

using excel

we have

SUMMARY OUTPUT
Regression Statistics
Multiple R	0.74754
R Square	0.558816
Adjusted R Square	0.382343
Standard Error	6.694215
Observations	15
ANOVA
	df	SS	MS	F	Significance F
Regression	4	567.6082	141.9021	3.166572	0.063417
Residual	10	448.1251	44.81251
Total	14	1015.733
	Coefficients	Standard Error	t Stat	P-value	Lower 95%	Upper 95%
Intercept	-10.9785	53.60105	-0.20482	0.841823	-130.409	108.4521
Smoker	14.17414	4.438576	3.193399	0.009599	4.284379	24.06391
Alcohol Consumption	0.32048	3.892888	0.082324	0.936013	-8.35341	8.994374
Weight (in lbs)	-0.01833	0.067697	-0.27082	0.79204	-0.16917	0.132505
Height (in inches)	0.383214	0.861736	0.4447	0.666007	-1.53685	2.303281

For Smoker : Since p value of smoker is less than 0.05 so Smoking is good predictor for life expectancy .

For Alcohol Consumption: Since p value of Alcohol Consumption is greater than 0.05 so we concloude that Alcohol consumption is not significant and not good predictor.

For Weight : Since p value of Weight is greater than 0.05 so we conclude that weight is not good predictor of life expectancy .

For Height: Since p value of Height is greater than 0.05 so we conclude that height is not good predictor of life expectancy .